CS Series

Advanced AI and HPC Cluster Computing

Contact Us

Get to the Solution Faster with Cray HPC Cluster Systems

Understanding the need for supercomputer-grade systems in an industry-standard form factor, we developed the Cray® CS™ series of cluster supercomputers. These HPC cluster systems feature the latest in processing, networking and cooling technologies, are highly customizable, and are expressly designed to handle the most demanding range of science, research, simulation, analytics and machine learning uses.

All CS components have been carefully selected for highest performance, optimized and integrated to create a powerful computing environment. Flexible node configurations featuring the latest processor and interconnect technologies mean you can tailor a cluster system to your specific needs — from an all-purpose HPC cluster to an accelerated system optimized to tackle AI training and inference workloads.

Beyond technology, Cray CS cluster users benefit from the depth of our 40+ years of supercomputing experience and the breadth of our partner ecosystem. From design and manufacturing to site engineering, system installation, optimization, ongoing management and support, CS series cluster supercomputers and software solutions are delivered with the same exacting standards and scalability as our legendary XC series supercomputing systems.

Featured Resources

Cray Accel AI Deep Learning Solutions

We give you the tools and support to guide you to success whether you’re just starting out or ready for production AI.

Cray CS500 Cluster Supercomputer Brochure

Cray CS500 cluster supercomputers are industry-standards-based, highly customizable, easy to manage and designed to handle the broadest range of simulations and data-intensive workloads.

Cray CS-Storm GPU-Accelerated Cluster Supercomputer Brochure

Purpose-built for the most demanding machine learning workloads, Cray CS-Storm technology provides customers with a powerful, accelerator-optimized solution for moving AI applications from pilot to production.

See All Resources

CLUSTER SYSTEM FEATURES & BENEFITS

Flexible, Scalable Configurations

The latest Intel® Xeon® Scalable family and Intel® Xeon Phi™ processors

Accelerated computing with NVIDIA® Tesla® GPUs and Nallatech FPGAs

Multiple interconnect options including InfiniBand™, Intel® OmniPath and Ethernet

High-performance parallel file systems including the Cray® ClusterStor™ scale-out Lustre® solution and third-party network-attached file systems

Air- or liquid-cooled designs

Easy to Manage

Industry-leading Bright Cluster Manager for HPC

NVIDIA Data Center GPU Manager (DCGM)

NVIDIA GPU-enabled Docker container

Ready for AI and Extreme HPC

A comprehensive machine learning and deep learning software environment

Cray Programming Environment for extreme HPC performance

Designed, Built, Installed and Serviced by Cray

Designed and architected for optimal compute, storage and networking performance

Built and tested at Cray’s supercomputer manufacturing facility

Turnkey installation services with worldwide support and service

Cray® CS-Storm™

Cray’s CS-Storm accelerated GPU computing system is designed for organizations that need the performance and efficiency of today’s latest GPU and FPGA accelerators. Whether it’s used for traditional engineering applications or in the rapidly-evolving fields of machine learning and deep learning, accelerated computing is quickly becoming mainstream in places where high-performance computing has traditionally been handled by CPUs.

Learn more about the CS-Storm system

Cray® CS™ Series

The CS series of cluster systems is designed to offer the widest possible choice of configurations. The Cray CS500 cluster supercomputer is a highly scalable and modular platform based on the latest x86 processing and and accelerator technologies from Intel, AMD and NVIDIA. Industry-standard server nodes and components have been optimized for high-performance computing (HPC) and tightly integrated with a comprehensive HPC software stack, creating a unified system that excels at capacity and data-intensive workloads.

Learn more about the CS500 system

Cray® CS™ Series Software  

Cray® CS™ series cluster supercomputers leverage the best in open source and commercial software for a complete HPC and artificial intelligence (AI) environment. Additionally, the CS series includes support for the legendary Cray Programming Environment (compilers, optimized debuggers, performance measurement and analysis tools, and scientific and math libraries).

Cray Cluster Software Stack

Bright Computing for Complete Cluster Management

Cray has partnered with Bright Computing to deliver a complete management and productivity environment for Cray® CS500™ and CS-Storm™ systems. The Bright Cluster Manager for HPC enables you to manage clustered infrastructure components including Ethernet and IB switches, PDUs, IP networks, primary and secondary head nodes (with automatic or manual failover), compute and login nodes, racks, chassis, processors, accelerators and file servers.

Bright Cluster Manager for HPC Capabilities:

  • Simplifies network, server, cluster fabric and accelerator management
  • Provisions individual nodes or complete clusters from bare metal
  • Integrates with leading workload managers: SLURM, Open Grid, Torque (built-in Moab/Maui), Altair PBS Professional®, and IBM Spectrum LSF
  • Facilitates automatic head node and workload manager failover, preventing cluster downtime
  • Offers scalable fast diskless booting for large node count systems
  • Manages heterogeneous nodes with different software stacks
  • Monitors a comprehensive set of hardware and software metrics that can be visualized and analyzed
  • Repurposes servers to accommodate fluctuating workloads on the fly

Deep Learning Extension for Bright Cluster Manager

For AI environments, Cray has partnered with Bright Computing to bring a comprehensive machine and deep learning environment to the CS series, making the latest in machine learning frameworks, libraries, GPU infrastructure and user-access tools available to Cray users. The Deep Learning Extension for Bright Cluster Manager enables complex machine and deep learning workloads and includes:

  • Machine and deep learning frameworks: Caffe, Cafe2, TensorFlow, MXnet, Microsoft Cognitive Toolkit, Torch, Theano and others
  • Machine learning libraries: the most popular machine learning libraries to help you access datasets, including MLPython, NVIDIA CUDA Deep Neural Network library (cuDNN) and Deep Learning GPU Training System (DIGITS)
  • Supporting infrastructure: over 400 MB of Python modules that support the machine learning packages, plus NVIDIA hardware drivers, CUDA (parallel computing platform API) drivers, CUB (CUDA building blocks), and NCCL (library of standard collective communication routines)

Product Literature

Cray Accel AI Deep Learning Solutions

We give you the tools and support to guide you to success whether you’re just starting out or ready for production AI.

Cray CS500 Cluster Supercomputer Brochure

Cray CS500 cluster supercomputers are industry-standards-based, highly customizable, easy to manage and designed to handle the broadest range of simulations and data-intensive workloads.

Cray CS-Storm GPU-Accelerated Cluster Supercomputer Brochure

Purpose-built for the most demanding machine learning workloads, Cray CS-Storm technology provides customers with a powerful, accelerator-optimized solution for moving AI applications from pilot to production.

White Papers and Briefs

Cray and NVIDIA: Artificial Intelligence at Any Size

Artificial intelligence and, in particular, deep learning are rapidly transforming entire industries and scientific disciplines. We give you the tools and support to guide you to success whether you’re just starting out or ready for production AI.

Blog

Cray Blog - Computing

Commentary and insight from Cray’s top experts and customers